1. Field of the Invention
The invention relates to the field of computer systems, and, more particularly, to a system to improve numerical conversion in such.
2. Description of Background
The Institute of Electrical and Electronics Engineers 754 floating standard definition of Decimal Floating-Points may have prompted the need for instructions(s) that convert between two different types of floating points (i.e., binary, hexadecimal, and decimal). One instruction called Perform Floating-Point operations (“PFPO”) may do such conversions. The PFPO instruction may attempt to convert different formats (short, long, or extended) of a floating-point data type to a specific format of a different floating-point type. For example, a short Binary Floating-Point (“BFP”) number may be converted to a short, long, or extended Decimal Floating-Point (“DFP”) number.
Hexadecimal is base 16 and has a fixed exponent length. Binary is base 2, and has more precision than hexadecimal. Binary's short, long, and extended formats have different exponent length. Decimal is base 10 and has the highest precision. Decimal suites well with financial applications and has recently been implemented in hardware (fast).